An industrial supplier cut wasted search spend by a third by structuring for intent

Representative examplesEvery case study in this library is an illustrative composite of the kind of engagement we deliver — written to show our method and standards, not to name clients.

Where they started

An industrial-supplies distributor serves trades and facilities teams from a warehouse operation that has run for two decades — counter sales, phone orders, and a web shop sitting on top of an ERP that tracks SKU margins precisely. Marketing spend was modest and concentrated in Google Search, managed for years as one campaign that 'worked'. The operations director reviewed cost per lead weekly from the finance side, and the account had accumulated years of settings from whoever touched it last — no documented structure, no owner of the search-terms report, no shared definition of a qualified enquiry.

What it was costing

One combined Search campaign served everything from 'hydraulic hose fittings' (buyers) to 'how hydraulic hoses work' (students). The budget spent evenly across both, the search-terms report had never been mined, and the operations director saw cost per lead triple over a year without knowing why.

What they could see

• Cost per lead had roughly tripled over the year while lead quality felt no better — the director's weekly check kept worsening with no explanation.
• The search-terms report showed a steady stream of 'how does' and 'what is' queries the account had been paying for, sometimes for months.
• Phone enquiries from ads often hit a line nobody staffed after hours, yet those calls sat in the conversion column feeding the bidding.
• A reseller's ads appeared above theirs whenever anyone searched the company's own name.
• High-margin product lines had no visibility advantage; the campaign treated a fitting and a full system as the same click.

The constraints we worked inside

• The director checked cost per lead weekly and was ready to cut the channel — the fix had to show in one quarter.
• SKU margins varied 8×; bid strategy had to respect product economics, not just conversion volume.
• A reseller competed on the brand terms — brand defense needed structure, not just a higher bid.

What had been tried before

The operations director had an intern export the search-terms report quarterly and add obvious junk phrases as negatives.

A quarterly sweep couldn't keep pace with query drift, and without intent-based campaign structure the research queries kept winning auctions anyway.

The account's bidding had been switched to maximize conversions the previous year on a predecessor's advice.

Smart bidding optimized toward the inflated conversion count, spending more to chase phone-call conversions nobody answered — the target itself was wrong.

Daily budget caps were lowered repeatedly to hold the monthly spend steady.

Caps rationed good queries along with junk; the mix never changed, so cost per lead kept climbing underneath the cap.

What we proposed

We proposed rebuilding the account around intent rather than keywords. Buyer-intent themes got their own campaigns with budgets tied to SKU margin tiers; research queries were fenced into a separate low-budget theme; and the search-terms report became a weekly mining routine feeding negatives and new ad groups. Before any of that, conversion tracking was rebuilt and verified against actual orders — restructuring on top of a broken conversion signal would just reorganize the error. The brand-defense question got its own exact-match structure rather than a higher blanket bid.

Just as important is what we ruled out, and why:

• Consolidating everything into one smart-bidding campaign — One campaign optimizes toward one conversion count; with SKU margins spread across a multiple wide enough to make blended targets meaningless, and research queries in the same auctions, automation would keep funding the junk.
• Broadening to Performance Max — PMax hides query-level control, and the core problem was query control — a weekly negatives routine can't run where search terms aren't visible.
• Outbidding the reseller across the account — The reseller problem was structural — brand defense needed its own exact-match campaign, not a higher blanket bid that raises costs on every query.

How the work ran

Delivered by the growth pod — paid search specialist over 6 weeks, with working increments reviewed with the client every week.

The stack, and the reasoning

Google Ads (Search)

Search kept the query-level control the restructure depended on — intent themes, negatives, and brand defense all live in visibility other campaign types don't offer.

Enhanced conversions + Consent Mode

Cost per lead had tripled on numbers nobody verified; enhanced conversions tied reports to real orders while consent handling kept the tracking compliant.

Negative keyword mining

The search-terms report was the account's largest untapped asset — a weekly routine turned it into the primary defense against research-query spend.

Feed-based Shopping

Shopping coverage ran from the product feed so high-margin SKUs could win product-listing slots the restructured search themes couldn't reach.

What went wrong

How we worked together

Cadence

A Tuesday call with the operations director, thirty minutes, screen-shared straight into the account — plus a one-page weekly summary of what the query mine added or removed.

Client side

The operations director owned decisions and the counter manager fed real enquiry examples; nobody else needed to attend, which suited a lean office.

Decisions

Calls got made on the Tuesday call; anything needing ERP data went to a short list the director answered by Friday.

They provided

ERP margin exports, phone-line answering hours in writing, and read access to the order history used for conversion verification.

What changed

The headline: cost per qualified enquiry over the quarter versus the prior one, on verified conversion tracking — −33%, read from Ads platform reconciled against CRM orders. A second check: return on ad spend for the high-margin sku campaign at 1.9×.

The director's weekly check now starts from a number he trusts, and the query mine runs whether we're watching or not. The counter staff noticed the difference first — the enquiries coming in name products, not homework topics. The brand-defense structure settled the reseller question without a phone call or a legal letter. The account has a documented structure for the first time, so the next person who touches it inherits a map instead of archaeology.

The result was read from Ads platform reconciled against CRM orders against the pre-engagement baseline over the stated window, with a guardrail check on return on ad spend for the high-margin sku campaign. Where platform-reported numbers and business outcomes differ, this record says which layer it is quoting.

What they own now

• The intent-tiered campaign structure with margin mappings documented in the account.
• A weekly query-mine runbook — what to negative, what to promote, what to leave.
• Verified conversion tracking with consent settings, reconciled against CRM orders.
• The exact-match brand-defense campaign with its cost-per-defense ceiling documented.
• A one-page margin-tier reference the director can re-map when the ERP export changes.